1. Field of the Invention
Certain embodiments of the present invention are generally directed at receivers that may, for example, be used for cellular system applications. Certain other embodiments of the present invention are generally directed at methods for operating such receivers.
2. Description of the Related Art
As personal, mobile, wireless communication devices have become more and more prevalent, low-cost, low-power receivers that may be implemented in such devices have been developed. Such receivers are generally preferred to have high degrees of sensitivity and linearity, particularly when used in receiver front-end portions. Therefore, because it has been demonstrated that bipolar complimentary metal oxide semiconductor (BiCMOS) technology is capable of providing the sensitivity and linearity desired, BiCMOS technology is often integrated into receivers according to the related art.
Using BiCMOS technology, the requirements of at least two mobile wireless communications systems, global system for mobile communications (GSM) and personal communications systems (PCS), have been met. Unfortunately, BiCMOS technology requires that a high number of masks be used when manufacturing receivers. Hence, the costs and complexity of manufacturing BiCMOS-based receivers are relatively high.
As wireless communication technology has continued to evolve, CMOS-based receivers have been studied as an alternative to the above-discussed BiCMOS technology. Using CMOS technology, relatively low-cost receivers may be produced. However, the noise of currently-available CMOS-based receivers is relatively high, and the linearity provided by such receivers still leaves much to be desired.
FIG. 1 illustrates a circuit diagram of a portion of CMOS-based single-balanced mixer 100 which may be included in a CMOS-based receiver according to the related art. Mixer 100 includes, in first transistor device 140, first local oscillator 110 and, in second transistor device 150, second local oscillator 120. Because neither oscillator 110 nor 120 is an ideal current source, noise component 130 is also included in the circuit diagram. Component 130 represents the equivalent noise voltage contributed from oscillator 110 and gate resistance thermal noise of transistor device 140. In other words, component 130 represents a combination of all nonlinear noise components associated with device 140.
FIG. 2A, in graph 200 contained therein, illustrates the differential voltage 205 that flows into single balanced mixer 100 over time (t) from local oscillators 110 and 120 when in operation. Also illustrated as a dashed line in graph 200 is an oscillating noise voltage (Vn) that modulates the local oscillator signals and that is present due to the fact that local oscillators 110 and 120 are not ideal sources.
Graph 210, in FIG. 2B, illustrates the combined differential output mixer current i0 that flows out of mixer 100 illustrated in FIG. 1. As shown in graph 210, the output current i0 takes the form of a step function that switches from a positive value (+I) equal to the maximum magnitude of the oscillator current to a negative value (−I) that is also equal in amplitude to the maximum magnitude of the oscillator current.
Current io may be decomposed into two components. The first component is an ideal differential output mixer current that is free of flicker noise and that has a 50% duty cycle. This first component, though not illustrated in FIGS. 2A-2C, takes the form of a step function that switches between upper and lower plateaus when differential voltage 205 crosses the t-axis in FIG. 2A.
Graph 220, illustrated in FIG. 2C, illustrates the second component of current io. This second component includes current spikes that represent flicker noise when mixer 100 is in operation. The current spikes are inherently present due to the configuration of mixer 100, and are caused by the leakage current that appears upon circuit switching.
The flicker noise illustrated in graph 220 leads to an offset of the above-discussed ideal differential output mixer current step function. In other words, without flicker noise, the steps in the step function illustrated in graph 210 would coincide exactly in time with when voltage swing 105 switched from a positive value to a negative value in graph 200. However, because flicker noise is superimposed on an ideal step function to represent the actual behavior of mixer 100 according to the related art, the positive or negative current spikes that represent flicker noise either increase or decrease the width of the steps in the ideal step function, thereby causing an offset between the edge of a step and the time at which voltage swing 105 switches from a positive value to a negative value. This offset reduces the sensitivity and linearity of currently-available CMOS-based wireless communication devices.
At least in view of the above, what is needed are new CMOS-based devices that may be used in wireless communication devices. Such devices should have higher sensitivity, lower noise, and higher degrees of linearity than currently available systems. What is also needed are methods for manufacturing and operating such devices.